直播预告 | ICLR专场四-白红宇

直播预告 | ICLR专场四

阅读量：2242 次

发布时间：2019-05-09

本文共 5222 字，大约阅读时间需要 17 分钟。

点击蓝字

关注我们

AI TIME欢迎每一位AI爱好者的加入！

6月2日晚7:30-9:00

AI TIME特别邀请了三位优秀的讲者跟大家共同开启ICLR专场四！

哔哩哔哩直播通道

扫码关注AITIME哔哩哔哩官方账号

观看直播

链接：https://live.bilibili.com/21813994

★ 邀请嘉宾 ★

瞿锰：魁北克人工智能研究院（Mila）三年级博士生，导师为Dr. Jian Tang。主要研究兴趣为图机器学习、概率模型、自然语言处理。现阶段的研究方向为结合深度学习与统计关系学习的知识推理。参与提出了图嵌入算法LINE，相关论文是WWW 2015被引用次数最高的论文。（个人主页：https://mnqu.github.io/）

报告题目：

基于逻辑规则学习的知识图谱推理

摘要：

We study learning logic rules for reasoning on knowledge graphs. Logic rules provide interpretable explanations when used for prediction as well as being able to generalize to other tasks, and hence are critical to learn. Existing methods either suffer from the problem of searching in a large search space (e.g., neural logic programming) or ineffective optimization due to sparse rewards (e.g., techniques based on reinforcement learning). To address these limitations, we propose a probabilistic model called RNNLogic. RNNLogic treats logic rules as a latent variable, and simultaneously trains a rule generator as well as a reasoning predictor with logic rules. We develop an EM-based algorithm for optimization. In each iteration, the reasoning predictor is updated to explore some generated logic rules for reasoning. Then in the E-step, we select a set of high-quality rules from all generated rules with both the rule generator and reasoning predictor via posterior inference; and in the M-step, the rule generator is updated with the rules selected in the E-step. Experiments on four datasets prove the effectiveness of RNNLogic.

论文标题：

RNNLogic: Learning Logic Rules for Reasoning on Knowledge Graphs

论文链接：

https://openreview.net/forum?id=tGZu6DlbreV

李恬：卡内基梅隆大学计算机科学系的博士生，导师是Virginia Smith。主要研究兴趣是分布式优化、联邦学习和数据密集型系统。在加入CMU之前，在北京大学获得计算机科学和经济学学士学位。（个人主页：https://www.cs.cmu.edu/~litian/）

报告题目：

Tilted Empirical Risk Minimization

摘要：

Empirical risk minimization (ERM) is typically designed to perform well on the average loss, which can result in estimators that are sensitive to outliers, generalize poorly, or treat subgroups unfairly. While many methods aim to address these problems individually, in this work, we explore them through a unified framework---tilted empirical risk minimization (TERM). In particular, we show that it is possible to flexibly tune the impact of individual losses through a straightforward extension to ERM using a hyperparameter called the tilt. We provide several interpretations of the resulting framework: We show that TERM can increase or decrease the influence of outliers, respectively, to enable fairness or robustness; has variance-reduction properties that can benefit generalization; and can be viewed as a smooth approximation to a superquantile method. We develop batch and stochastic first-order optimization methods for solving TERM, and show that the problem can be efficiently solved relative to common alternatives. Finally, we demonstrate that TERM can be used for a multitude of applications, such as enforcing fairness between subgroups, mitigating the effect of outliers, and handling class imbalance. TERM is not only competitive with existing solutions tailored to these individual problems, but can also enable entirely new applications, such as simultaneously addressing outliers and promoting fairness.

论文标题：

Tilted Empirical Risk Minimization

论文链接：

https://openreview.net/forum?id=K5YasWXZT3O

齐浩之：美国加州大学伯克利分校博士生，导师是马毅教授和马利克教授。从香港科技大学(科大)取得学士学位，导师是邓志强教授。主要研究兴趣是动态建模和基于学习的机器人操作。（个人主页：https://haozhi.io）

报告题目：

Learning Long-term Visual Dynamics with Region Proposal Interaction Networks

摘要：

Learning long-term dynamics models is the key to understanding physical common sense. Most existing approaches on learning dynamics from visual input sidestep long-term predictions by resorting to rapid re-planning with short-term models. This not only requires such models to be super accurate but also limits them only to tasks where an agent can continuously obtain feedback and take action at each step until completion. In this paper, we aim to leverage the ideas from success stories in visual recognition tasks to build object representations that can capture inter-object and object-environment interactions over a long-range. To this end, we propose Region Proposal Interaction Networks (RPIN), which reason about each object's trajectory in a latent region-proposal feature space. Thanks to the simple yet effective object representation, our approach outperforms prior methods by a significant margin both in terms of prediction quality and their ability to plan for downstream tasks, and also generalize well to novel environments. Code, pre-trained models, and more visualization results are available at this https URL.

论文标题：

Learning Long-term Visual Dynamics with Region Proposal Interaction Networks

论文链接：

https://arxiv.org/abs/2008.02265

直播结束后我们会邀请讲者在微信群中与大家答疑交流，请添加“AI TIME小助手（微信号：AITIME_HY）”，回复“iclr”，将拉您进“ICLR会议交流群”！